From 073e5d6642b1cf4becb21c696fe563ae6aa8d53d Mon Sep 17 00:00:00 2001
From: Vitor Mattos <1079143+vitormattos@users.noreply.github.com>
Date: Fri, 24 Apr 2026 00:15:17 -0300
Subject: [PATCH] fix: accept multi-space xref subsection entries

Signed-off-by: Vitor Mattos <1079143+vitormattos@users.noreply.github.com>
---
 .../PullRequestXrefSubsectionMultipleSpaces.pdf  | Bin 0 -> 2541 bytes
 src/Smalot/PdfParser/RawData/RawDataParser.php   |   2 +-
 .../Integration/DocumentIssueFocusTest.php       |   7 +++++++
 3 files changed, 8 insertions(+), 1 deletion(-)
 create mode 100644 samples/bugs/PullRequestXrefSubsectionMultipleSpaces.pdf

diff --git a/samples/bugs/PullRequestXrefSubsectionMultipleSpaces.pdf b/samples/bugs/PullRequestXrefSubsectionMultipleSpaces.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..508c197476937691574c8152d80bc4bdf6dbec27
GIT binary patch
literal 2541
zcmbtWU2fwx5N^>o0lmY(ivS4<nWQY)lHhpZST?q3ya}8XEfDMjEzyp;wj_{rqvR}m
zr+w%R+KY53`A6$`fh{WdK^hLf`DQrZ&}X;N#Aa>RI{W*d&wp8r0fl`2ht=;}WRkA5
zkU%CYu0<r`JP{VT7THoif^Q85meNx2wN-xl;X8c3w?vj~P55R|Sf>GZ|6B+lQ=xgn
zHMhv!X09vK@6+o8t{*gz`)|vF#~+1;xmc#zWvc~ga@qRLo6>195|8P-7b)&u-rvVB
zALAfty&YIqzYNQDvBoe^u2-1~%gYu=HX_7#V<xSE1$~(;!dW!g`ry}P>rv|>B;@Jo
zsr}Sx=W<C{5Cnue#Bpr&u+?*>d1+_rje)2!BB5fL7COx{FqnLvZ?qv=2QQKZH<TOL
zJWbaXx>+NcB56`lLcrQPEV$GI)+fzRjJIEp^K6mJHP2!(xZ1llt%e;jbQ<*j1LZy6
zOtq`pwc~tk=n&TKG{}axGEX+Kkb{o>PNd66t<LYK7l2I#yCq=lUYoXiaNei^LfJac
zSIPiiR4E*={ttOP`qSh~ULbk8BaeRJ2)kG3@*a7d;J;n{Ci=%{)nV<8B0~KP97=tg
zy7;@N-Oy$D`-#$!(mpxz-0z;|NxFFcmfa1{uT-yY2kiKPg#mXG!H>PfP_vgUS!u|6
zE#=orzU51t2e({`OoM|$pIKxi^WrKm;eq0}7kJ>i&Ib!cpQQYReI``Ck+D#K^9BbB
z8ReNqI0e-pni<=6l$Zd!!BkHY6JJ^6mo!oEVCc;-(kR~`8`jv!43+)PRIdb_i?R^0
z7746{QV0_Q4kLvoqvA)w6Cp!T8fBPfOS2i5Alt0xI1YK1faKXyK$dGz0>z5Qo51`T
zI4JbT01Bs;XkYBB;~94NCB>f*)bBM>+8}BczQ$7uwOQ#9aG0!4!}?#slJ}y-I+)y>
zRWx2lG13eZ#YqDEG;r?}aGk{Abd1vxuFi`z2FgyOc2A+!IUG=p(-CUC;u|HPYctz@
zg}ke2oqSFFfMyQi2o2{H8G;rH4|0=g?1NO{q36=B0@#`|+mzgH)W&Tx_slIRar-u(
zYmV=)s&2OIBUe}6%|@@%%(U;WVYfKjPN{I}_d^2BySbLY2rtrAzBGM?TG)wj2M$}R
zH-*TCb#K=?!7Dx0C>FEpnm^-(1>~;k+?}SE0w)UI*1syTLo+I_uWqQW^=4a@QY<VA
zz+8)NALc;U^Ew_Ze%#tv&_zP#foyMM0V?;%M!T5#k<CBh*Xv<Lj{LeO{JIod;mFTp
z*uqCP_k_*yP(ok&QOaKNt2}DIeZG_ijeNDlJ`%F}V|ACl2vaLE%W;I5W5wa{a|93l
z$aRO&(1{|)A9u#XV8VPqa-vBTb=>aIbq3HMIpe_f9XgtLkvE=n>DAb|>M?iX)1KE0
Yt_B~>OsR%fbu12JzIApszL{A60{WxZ%m4rY

literal 0
HcmV?d00001

diff --git a/src/Smalot/PdfParser/RawData/RawDataParser.php b/src/Smalot/PdfParser/RawData/RawDataParser.php
index ec8d01e5..0df21551 100644
--- a/src/Smalot/PdfParser/RawData/RawDataParser.php
+++ b/src/Smalot/PdfParser/RawData/RawDataParser.php
@@ -169,7 +169,7 @@ protected function decodeXref(string $pdfData, int $startxref, array $xref = [],
         // initialize object number
         $obj_num = 0;
         // search for cross-reference entries or subsection
-        while (preg_match('/([0-9]+)[\x20]([0-9]+)[\x20]?([nf]?)(\r\n|[\x20]?[\r\n])/', $pdfData, $matches, \PREG_OFFSET_CAPTURE, $offset) > 0) {
+        while (preg_match('/([0-9]+)[\x20]+([0-9]+)[\x20]*([nf]?)(\r\n|[\x20]*[\r\n])/', $pdfData, $matches, \PREG_OFFSET_CAPTURE, $offset) > 0) {
             if ($matches[0][1] != $offset) {
                 // we are on another section
                 break;
diff --git a/tests/PHPUnit/Integration/DocumentIssueFocusTest.php b/tests/PHPUnit/Integration/DocumentIssueFocusTest.php
index 7c7fe7e6..3d5d3de5 100644
--- a/tests/PHPUnit/Integration/DocumentIssueFocusTest.php
+++ b/tests/PHPUnit/Integration/DocumentIssueFocusTest.php
@@ -111,4 +111,11 @@ public function testPDFDocEncodingDecode(): void
         $testSubject = '•†‡…—–ƒ⁄‹›−‰„“”‘’‚™ŁŒŠŸŽıłœšž';
         self::assertStringContainsString($testSubject, $details['Subject']);
     }
+
+    public function testParseFileWithXrefSubsectionHavingMultipleSpaces(): void
+    {
+        $document = (new Parser())->parseFile($this->rootDir.'/samples/bugs/PullRequestXrefSubsectionMultipleSpaces.pdf');
+
+        self::assertCount(1, $document->getPages());
+    }
 }